請用此 Handle URI 來引用此文件：
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/86500完整後設資料紀錄
| DC 欄位 | 值 | 語言 |
|---|---|---|
| dc.contributor.advisor | 魏宏宇(Hung-Yu Wei) | |
| dc.contributor.author | Ya-Ting Yang | en |
| dc.contributor.author | 楊雅婷 | zh_TW |
| dc.date.accessioned | 2023-03-19T23:59:29Z | - |
| dc.date.copyright | 2022-08-18 | |
| dc.date.issued | 2022 | |
| dc.date.submitted | 2022-08-15 | |
| dc.identifier.citation | [1] Tsung-Yi Lin, Michael Maire, Serge Belongie, Lubomir Bourdev, Ross Girshick, James Hays, Pietro Perona, Deva Ramanan, C. Lawrence Zitnick, and Piotr Dollár. Microsoft coco: Common objects in context, 2015. [2] Ruhul Amin Khalil, Nasir Saeed, Mudassir Masood, Yasaman Moradi Fard, Mohamed-Slim Alouini, and Tareq Y. Al-Naffouri. Deep learning in the industrial internet of things: Potentials, challenges, and emerging applications. IEEE Internet of Things Journal, 8(14):11016–11040, 2021. [3] Armstrong Aboah. A vision-based system for traffic anomaly detection using deep learning and decision trees. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, pages 4207–4212, June 2021. [4] Jiale Cao, Yanwei Pang, Jin Xie, Fahad Shahbaz Khan, and Ling Shao. From hand-crafted to deep features for pedestrian detection: A survey. IEEE Transactions on Pattern Analysis and Machine Intelligence, page 1–1, 2021. [5] Sweta Bhattacharya, Siva Rama Krishnan Somayaji, Thippa Reddy Gadekallu, Mamoun Alazab, and Praveen Kumar Reddy Maddikunta. A review on deep learning for future smart cities. Internet Technology Letters, n/a(n/a):e187. [6] Yongbin Gao, Xuehao Xiang, Naixue Xiong, Bo Huang, HyoJong Lee, Rad Alrifai, Xiaoyan Jiang, and Zhijun Fang. Human action monitoring for healthcare based on deep learning. IEEE Access, 6:52277–52285, 2018. [7] Trang Pham, Truyen Tran, Dinh Phung,and Svetha Venkatesh. Predicting healthcare trajectories from medical records: A deep learning approach. Journal of Biomedical Informatics, 69, 04 2017. [8] Guruh Fajar Shidik, Edi Noersasongko, Adhitya Nugraha, Pulung Nurtantio Andono, Jumanto Jumanto, and Edi Jaya Kusuma. A systematic review of intelligence video surveillance: Trends, techniques, frameworks, and datasets. IEEE Access, 7:170457–170473, 2019. [9] Xiaoqiang Ma, Tai Yao, Menglan Hu, Yan Dong, Wei Liu, Fangxin Wang, and Jiangchuan Liu. A survey on deep learning empowered iot applications. IEEE Access, 7:181721–181732, 2019. [10] Niall O’Mahony, Sean Campbell, Anderson Carvalho, Suman Harapanahalli, Gustavo Velasco Hernandez, Lenka Krpalkova, Daniel Riordan, and Joseph Walsh. Deep learning vs. traditional computer vision. In Kohei Arai and Supriya Kapoor, editors, Advances in Computer Vision, pages 128–144, Cham, 2020. Springer International Publishing. [11] Zhi Zhou, Xu Chen, En Li, Liekang Zeng, Ke Luo, and Junshan Zhang. Edge intelligence: Paving the last mile of artificial intelligence with edge computing. Proceedings of the IEEE, 107(8):1738–1762, 2019. [12] Te-Yi Kan, Yao Chiang, and Hung-Yu Wei. Task offloading and resource allocation in mobile-edge computing system. In 2018 27th Wireless and Optical Communication Conference (WOCC), pages 1–4, 2018. [13] Hongzhi Guo, Jiajia Liu, Ju Ren, and Yanning Zhang. Intelligent task offloading in vehicular edge computing networks. IEEE Wireless Communications, 27(4):126– 132, 2020. [14] Yishi Zhu, Bomin Mao, Yuichi Kawamoto, and Nei Kato. Intelligent reflecting surface-aided vehicular networks toward 6g: Vision, proposal, and future directions. IEEE Vehicular Technology Magazine, 16(4):48–56, 2021. [15] Hongzhi Guo and Jiajia Liu. Uav-enhanced intelligent offloading for internet of things at the edge. IEEE Transactions on Industrial Informatics, 16(4):2737–2746, 2020. [16] Bomin Mao, Fengxiao Tang, Yuichi Kawamoto, and Nei Kato. Optimizing computation offloading in satellite-uav-served 6g iot: A deep learning approach. IEEE Network, 35(4):102–108, 2021. [17] Chien-Chun Hung, Ganesh Ananthanarayanan, Peter Bodik, Leana Golubchik, Min-lan Yu, Paramvir Bahl, and Matthai Philipose. Videoedge: Processing camera streams using hierarchical clusters. In 2018 IEEE/ACM Symposium on Edge Computing (SEC), pages 115–131, 2018. [18] Yi Zhang, Jiun-Hao Liu, Chih-Yu Wang, and Hung-Yu Wei. Decomposable intelligence on cloud-edge iot framework for live video analytics. IEEE Internet of Things Journal, 7(9):8860–8873, 2020. [19] Zhuoran Zhao, Kamyar Mirzazad Barijough, and Andreas Gerstlauer. Deepthings: Distributed adaptive deep learning inference on resource-constrained iot edge clusters. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 37(11):2348–2359, 2018. [20] Ramyad Hadidi, Jiashen Cao, Michael S. Ryoo, and Hyesoon Kim. Toward collaborative inferencing of deep neural networks on internet-of-things devices. IEEE Internet of Things Journal, 7(6):4950–4960, 2020. [21] Liekang Zeng, Xu Chen, Zhi Zhou, Lei Yang, and Junshan Zhang. Coedge: Cooperative dnn inference with adaptive workload partitioning over heterogeneous edge devices, 2020. [22] Emre Kilcioglu, Hamed Mirghasemi, Ivan Stupia, and Luc Vandendorpe. An energy-efficient fine-grained deep neural network partitioning scheme for wireless collaborative fog computing. IEEE Access, 9:79611–79627, 2021. [23] Jingyue Zhou, Yihuai Wang, Kaoru Ota, and Mianxiong Dong. Aaiot: Accelerating artificial intelligence in iot systems. IEEE Wireless Communications Letters, 8(3):825–828, 2019. [24] En Li, Liekang Zeng, Zhi Zhou, and Xu Chen. Edge ai: On-demand accelerating deep neural network inference via edge computing, 2019. [25] Liekang Zeng, En Li, Zhi Zhou, and Xu Chen. Boomerang: On-demand cooperative deep neural network inference for edge intelligence on the industrial internet of things. IEEE Network, 33(5):96–103, 2019. [26] Chuang Hu, Wei Bao, Dan Wang, and Fengming Liu.Dynamic adaptive dnn surgery for inference acceleration on the edge. In IEEE INFOCOM 2019 - IEEE Conference on Computer Communications, pages 1423–1431, 2019. [27] Chunwei Xia, Jiacheng Zhao, Huimin Cui, Xiaobing Feng, and Jingling Xue. Dnntune: Automatic benchmarking dnn models for mobile-cloud computing. ACM Trans. Archit. Code Optim., 16(4), December 2019. [28] Yutao Huang, Feng Wang, Fangxin Wang, and Jiangchuan Liu. Deepar: A hybrid device-edge-cloud execution framework for mobile deep learning applications. In IEEE INFOCOM 2019 - IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS), pages 892–897, 2019. [29] Wenchen He, Shaoyong Guo,Song Guo, Xuesong Qiu, and Feng Qi.Joint dnn partition deployment and resource allocation for delay-sensitive deep learning inference in iot. IEEE Internet of Things Journal, 7(10):9241–9254, 2020. [30] Chongwu Dong, Sheng Hu, Xi Chen, and Wushao Wen. Joint optimization with dnn partitioning and resource allocation in mobile edge computing. IEEE Transactions on Network and Service Management, 18(4):3973–3986, 2021. [31] 3GPP. 5G System (5GS); Study on traffic characteristics and performance requirements for AI/ML model transfer. Technical Report (TR) 22.874, 3rd Generation Partnership Project (3GPP). [32] Zecheng He, Tianwei Zhang, and Ruby B Lee. Model inversion attacks against collaborative inference. In Proceedings of the 35th Annual Computer Security Applications Conference, pages 148–162. ACM, 2019. [33] Zecheng He, Tianwei Zhang, and Ruby B Lee.Attacking and protecting data privacy in edge-cloud collaborative inference systems. IEEE Internet of Things Journal, 2020. [34] Noam Nisan, Tim Roughgarden, Éva Tardos, and Vijay Vazirani. Algorithmic Game Theory. 01 2007. [35] Yang Zhang, Chonho Lee, Dusit Niyato, and Ping Wang. Auction approaches for resource allocation in wireless systems: A survey. IEEE Communications Surveys Tutorials, 15(3):1020–1041, 2013. [36] Houming Qiu, Kun Zhu, Nguyen Cong Luong, Changyan Yi, Dusit Niyato, and Dong In Kim. Applications of auction and mechanism design in edge computing: A survey, 2021. [37] William Vickrey. Counterspeculation, auctions, and competitive sealed tenders. The Journal of Finance, 16(1):8–37, 1961. [38] Edward Clarke. Multipart pricing of public goods. Public Choice, 11(1):17–33, 1971. [39] Theodore Groves. Incentives in teams. Econometrica, 41(4):617–631, 1973. [40] Pierre Baqué, François Fleuret, and Pascal V. Fua. Deep occlusion reasoning for multi-camera multi-target detection. 2017 IEEE International Conference on Com- puter Vision (ICCV), pages 271–279, 2017. [41] Tatjana Chavdarova and Francois Fleuret. Deep multi-camera people detection. In 2017 16th IEEE International Conference on Machine Learning and Applications (ICMLA), pages 848–853, 2017. [42] Zhu Han, Dusit Tao Niyato, Walid Saad, Tamer Bąar, and Are Hjrungnes. Game theory in wireless and communication networks: Theory, models, and applications. 2011. [43] Min Lin, Qiang Chen, and Shuicheng Yan. Network in network, 2014. [44] Karen Simonyan and Andrew Zisserman. Very deep convolutional networks for large-scale image recognition, 2015. [45] Alex Krizhevsky, Ilya Sutskever, and Geoffrey E Hinton. Imagenet classification with deep convolutional neural networks. In F. Pereira, C. J. C. Burges, L. Bottou, and K. Q. Weinberger, editors, Advances in Neural Information Processing Systems, volume 25. Curran Associates, Inc., 2012. [46] Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition, 2015. [47] Alejandro López-Cifuentes, Marcos Escudero-Viñolo, Jesús Bescós, and Pablo Carballeira. Semantic driven multi-camera pedestrian detection. ArXiv, abs/1812.10779, 2018. [48] J. Lima, R. Roberto, L. Figueiredo, F. Simoes, and V. Teichrieb. Generalizable multi-camera 3d pedestrian detection. In 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pages 1232–1240, Los Alamitos, CA, USA, jun 2021. IEEE Computer Society. [49] Ezgi Ozyilkan and Mikolaj Jankowski. Deep joint transmission-recognition for multi-view cameras. ArXiv, abs/2011.01902, 2020. [50] Walid Saad, Zhu Han, Merouane Debbah, Are Hjorungnes, and Tamer Basar. Coalitional game theory for communication networks. IEEE Signal Processing Magazine, 26(5):77–97, 2009. [51] Faheem Zafari, Kin K.Leung, Don Towsley, Prithwish Basu, Ananthram Swami,and Jian Li. Let’s share: A game-theoretic framework for resource sharing in mobile edge clouds. IEEE Transactions on Network and Service Management, 18(2):2107– 2122, 2021. [52] Quoc-Viet Pham, Hoang T. Nguyen, Zhu Han, and Won-Joo Hwang. Coalitional games for computation offloading in noma-enabled multi-access edge computing. IEEE Transactions on Vehicular Technology, 69(2):1982–1993, 2020. [53] Jizhe Zhou, Xing Zhang, and Wenbo Wang. Joint resource allocation and user association for heterogeneous services in multi-access edge computing networks. IEEE Access, 7:12272–12282, 2019. [54] 3GPP. Study on channel model for frequencies from 0.5 to 100 GHz. Technical Report (TR) 38.901, 3rd Generation Partnership Project (3GPP). [55] 3gpp. Study on new radio access technology Physical layer aspects. Technical Report (TR) 38.802, 3rd Generation Partnership Project (3GPP). [56] Alexey Bochkovskiy, Chien-Yao Wang, and Hong-Yuan Mark Liao. Yolov4: Optimal speed and accuracy of object detection, 2020. [57] Runwei Ding, Linhui Dai, Guangpeng Li, and Hong Liu. Tdd-net: A tiny defect detection network for printed circuit boards. CAAI Transactions on Intelligence Technology, 4, 04 2019. [58] Tim Roughgarden. Twenty Lectures on Algorithmic Game Theory. Cambridge University Press, 2016. [59] Ya-Ting Yang and Hung-Yu Wei. Edge computing and networking resource management for decomposable deep learning: An Auction-Based approach. In The 22nd Asia-Pacific Network Operations and Management Symposium) (APNOMS 2021), Tainan, Taiwan, September 2021. [60] Fei-Fei Li, Marco Andreeto, Marc’Aurelio Ranzato, and Pietro Perona. Caltech 101, Apr 2022. [61] Krzysztof R. Apt and Tadeusz Radzik. Stable partitions in coalitional games. ArXiv, abs/cs/0605132, 2006. | |
| dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/86500 | - |
| dc.description.abstract | 隨著各種物聯網 (IoT) 系統需求的快速增長，深度學習 (DL) 相關的應用也引起了廣泛關注。然而，在計算能力有限的物聯網系統上執行深度學習模型的推論 (inference) 非常具有挑戰性。與雲端計算相比，邊緣計算 (edge computing) 將資源部署在終端用戶附近以減少傳輸延遲並保留原始數據以減輕隱私洩漏的問題。但由於邊緣的資源還是有限的，用戶跟邊緣節點之間的關聯 (association) 決策、邊緣節點的資源分配、使用的深度學習模型參數選擇，以及設備配置等管理就變得至關重要。此外，3GPP 最近為下一代無線網絡討論的機器學習推論分割 (split-ML) 概念也為邊緣管理提供了更大的靈活性。在本文中，我們首先通過 Vickrey-Clarke-Groves 機制，在考慮物聯網設備偏好的同時研究準確率和每秒幀數 (FPS) 要求下的邊緣節點資源分配問題。然後，加入機器學習推論分割中的隱私問題和多視角檢測提供的偵測準確度增強，我們將研究擴大到物聯網設備與邊緣節點之間的關聯決策問題。我們使用賽局中形成聯盟 (coalition formation) 的演算法來啟發式地優化社會福利，並證明最終聯盟結構的收斂性和穩定性。模擬結果表明，這兩個問題中的設計旋鈕都是必不可少的，並且確實在不同的情景設置下為邊緣運算的管理帶來了更好的整體效用。 | zh_TW |
| dc.description.abstract | Deep learning (DL) applications have attracted significant attention with the rapidly growing demand for Internet of Things (IoT) systems. However, it is challenging to perform the inference of such resource-hungry DL models on the computationally limited IoT system. Compared to cloud computing, edge computing deploys resources near the end-users to reduce the transmission delay while retaining the raw data to mitigate privacy concerns. Since resources at the edge are limited, management like user association decisions, edge resources allocation, and device configuration with DL model parameter selection becomes essential. Moreover, the concept of split-ML recently discussed by 3GPP for the next-generation wireless networks also provides more flexibility for management at the edge. In this work, we first study the joint edge resources allocation problem under accuracy and FPS requirements while considering IoT devices' preferences by applying the Vickrey–Clarke–Groves-based method. Then, we enlarge the problem to make the association decision between IoT devices and the edge nodes, considering the privacy retain issue in split-ML and the enhancement provided by multi-view detection. We handle the association problem heuristically by a coalition formation-based algorithm and prove properties such as convergence and stability. The simulation results show that the design knobs in both problems are essential and indeed lead to better social welfare under different scenario settings. | en |
| dc.description.provenance | Made available in DSpace on 2023-03-19T23:59:29Z (GMT). No. of bitstreams: 1 U0001-1308202222421300.pdf: 11134325 bytes, checksum: 8c070b413bf9e65a91057403d048dc8e (MD5) Previous issue date: 2022 | en |
| dc.description.tableofcontents | 摘要 v Abstract vii 1 Introduction 1 1.1 Background and Motivation ........................ 1 1.2 Contributions ................................ 4 1.2.1 For system with an edge node and several IoT devices . . . . . . 4 1.2.2 For system with multiple edge nodes and several IoT devices . . . 5 2 Related Works 7 2.1 Split-ML .................................. 7 2.1.1 Model-Level Decomposition.................... 7 2.1.2 Parallel Decomposition....................... 7 2.1.3 Layer-Level Decomposition .................... 8 2.1.4 Joint Resource Allocation ..................... 9 2.2 Edge Camera Network ........................... 9 2.2.1 Multi-view Detection........................ 9 2.2.2 Coalition Formation ........................ 10 3 Edge-IoT Computing and Networking Resource Allocation for Decomposable Deep Learning Inference 13 3.1 System Model................................ 14 3.1.1 Networking Model ......................... 14 3.1.2 Computing Model ......................... 15 3.1.3 Configuration Table ........................ 15 3.2 Problem Formulation ............................ 18 3.2.1 The VCG mechanism........................ 18 3.2.2 Resource allocation for split DL image recognition . . . . . . . . 19 3.2.3 Resource allocation for split DL video recognition with FPS constraint................................ 23 3.2.4 Model Properties.......................... 26 3.3 Evaluation Methodology .......................... 29 3.3.1 Baseline for a single application: the weighted method . . . . . . 29 3.3.2 Baseline for multi-application: the ratio method . . . . . . . . . . 30 3.3.3 Baseline: offload all or none.................... 30 3.3.4 Parameters for evaluation ..................... 30 3.4 Performance Evaluation Results ...................... 33 3.4.1 Scenario: Single Application Type................. 33 3.4.2 Scenario: Multiple Application Types . . . . . . . . . . . . . . . 35 3.4.3 Impact of Device Preference on Allocated Resources . . . . . . . 35 3.4.4 Truthfulness of the Mechanism................... 36 3.4.5 Impact of different numbers of IoT devices . . . . . . . . . . . . 37 3.4.6 The trade-off between FPS requirement and accuracy . . . . . . . 38 3.5 Conclusions................................. 39 4 A Coalition Formation Approach for Privacy and Energy-Aware Split Deep Learning Inference in Edge Camera Network 47 4.1 System Model and Problem Formulation.................. 47 4.1.1 System Overview.......................... 48 4.1.2 DNN Partition and Camera Configuration . . . . . . . . . . . . . 50 4.1.3 Problem Formulation........................ 53 4.2 Coalition Formation Game ......................... 55 4.2.1 Coalition Formation Game Model ................. 55 4.2.2 Coalition Formation Game Based Algorithm . . . . . . . . . . . 57 4.2.3 Proposed Algorithm Analysis ................... 59 4.3 Simulation Results ............................. 61 4.3.1 Simulation Parameters ....................... 61 4.3.2 Baseline Schemes.......................... 62 4.3.3 Ablation Study ........................... 64 4.3.4 Impact of the overlapping view matrix. . . . . . . . . . . . . . . 65 4.3.5 Impact of the connection matrix .................. 66 4.3.6 Convergence and Stability of Proposed Method . . . . . . . . . . 68 4.4 Conclusions................................. 69 5 Conclusions 71 6 Appendix 73 6.1 Additional simulation results for section4.3 . . . . . . . . . . . . . . 73 6.1.1 Impact of the overlapping view matrix. . . . . . . . . . . . . . . 73 6.1.2 Impact of the connection matrix .................. 75 Bibliography 77 | |
| dc.language.iso | en | |
| dc.subject | 資源分配 | zh_TW |
| dc.subject | 賽局方法 | zh_TW |
| dc.subject | 資源分配 | zh_TW |
| dc.subject | 機器學習推論分割 | zh_TW |
| dc.subject | 邊緣運算 | zh_TW |
| dc.subject | 賽局方法 | zh_TW |
| dc.subject | 機器學習推論分割 | zh_TW |
| dc.subject | 邊緣運算 | zh_TW |
| dc.subject | Split-ML | en |
| dc.subject | Game Theory | en |
| dc.subject | Edge Computing | en |
| dc.subject | Game Theory | en |
| dc.subject | Split-ML | en |
| dc.subject | Edge Computing | en |
| dc.subject | Resource Allocation | en |
| dc.subject | Resource Allocation | en |
| dc.title | 基於賽局方法為物聯網的可拆分之深度神經網路推論在邊緣運算中進行資源管理 | zh_TW |
| dc.title | Game-Theoretic Resource Management for Decomposable DNN Inference in Edge-IoT Framework | en |
| dc.type | Thesis | |
| dc.date.schoolyear | 110-2 | |
| dc.description.degree | 碩士 | |
| dc.contributor.oralexamcommittee | 陸寶森(Peter B. Luh),張時中(Shi-Chung Chang),陳和麟(Ho-Lin Chen),王志宇(Chih-Yu Wang) | |
| dc.subject.keyword | 邊緣運算,機器學習推論分割,資源分配,賽局方法, | zh_TW |
| dc.subject.keyword | Edge Computing,Split-ML,Resource Allocation,Game Theory, | en |
| dc.relation.page | 84 | |
| dc.identifier.doi | 10.6342/NTU202202371 | |
| dc.rights.note | 同意授權(全球公開) | |
| dc.date.accepted | 2022-08-16 | |
| dc.contributor.author-college | 電機資訊學院 | zh_TW |
| dc.contributor.author-dept | 電信工程學研究所 | zh_TW |
| dc.date.embargo-lift | 2022-08-18 | - |
| 顯示於系所單位： | 電信工程學研究所 | |
文件中的檔案：
| 檔案 | 大小 | 格式 | |
|---|---|---|---|
| U0001-1308202222421300.pdf | 10.87 MB | Adobe PDF | 檢視/開啟 |
系統中的文件，除了特別指名其著作權條款之外，均受到著作權保護，並且保留所有的權利。